Loading
Why Your Team Still Does Manual Work Even When You Have Enough Software
·4 min read
Most businesses that have invested in software still run on a layer of manual effort underneath it. Someone copies a number from one system into another. A manager chases a colleague for a status update, that the system itself should already contain. A spreadsheet sits beside a platform that was supposed to replace it. If the tools are there, why does the manual work persist?
When organisations look for an explanation, software quality and training gaps come up most often. Both may be true in particular cases, but neither gets to the root of what is happening in most organisations. We think something structural is at work, and it shows up in the data with some consistency.
When a business acquires software, it acquires a system designed around a particular view of how work should flow. Over time, the actual work changes: new processes emerge, teams find shortcuts, customer requirements shift. The software stays roughly where it was. Whatever distance opens between what the tool expects and what the job requires gets filled with manual effort, usually by the person closest to the work, usually without anyone formally deciding that is how it should work.
84% of organisations have not redesigned jobs or workflows around AI, even as access to AI tools grew 50% in a single year. Adding tools without redesigning the work around them is the pattern that produces this accumulation. The software accumulates; the workflows do not keep pace.
Working days shaped more by friction than by the work itself follow from this. Workers lose the equivalent of 51 working days a year to technology friction, across a survey of 3,750 executives and employees in 14 countries. That covers the time spent navigating between systems, re-entering information, waiting for access and working around tools that do not fit the task.
Even where the software fits the work reasonably well, employees tend to reach past it. More than half of workers bypassed AI tools and completed tasks manually at least once in the past 30 days, and a further 33% had not used AI at all during that period, in the same survey. 61% of employees prefer familiar tools even when much better, company-approved alternatives exist, because switching systems carries a cognitive cost that outweighs the benefit of better options in the middle of a busy day.
A tool that requires more effort to use than the manual alternative loses to the manual alternative, regardless of what the tool is capable of on paper. That changes only when the tool is faster, more reliable and clearly sanctioned for the task.
Additionally, opinions about software varies between executives and employees. 88% of executives said employees had adequate tools; only 21% of workers agreed, in the same study. The people deciding which tools to buy are working from a different view of daily operations than the people using them.
Three conditions tend to reduce manual work: the tool fits the actual task, the person using it has a clear reason to trust it, and the workflow has been redesigned to route work through it. Automation helps most where the task is repetitive, the inputs are consistent and the cost of a manual error is real.
Document-heavy processes are a common candidate: executives estimate that employees still spend more than six hours each week on manual document-related tasks, with 41% putting the figure between 11 and 15 hours, in a survey of more than 1,300 enterprise leaders. Recovering that time requires redesigning the workflow around the tool, with clear ownership of what the system handles and what a person reviews.
For any persistent manual task, the obvious question is why there's no tool handling it already? And if there is a tool, is the workflow redesigned for the tool does not fit the actual task? Or is there a reality where the manual alternative is still faster for the person doing the work? Each question has a different remedy, and identifying which one applies precedes any useful automation decision.
[Workflow automation: how to automate repetitive business tasks with AI](https://prudos.ai/intelligence/workflow-automation-how-to-automate-repetitive-business-task-1scl) covers how to approach that redesign in practice.